- A survey among more than 1800 climate scientists confirms that there is widespread agreement that global warming is predominantly caused by human greenhouse gases.
- This consensus strengthens with increased expertise, as defined by the number of self-reported articles in the peer-reviewed literature.
- The main attribution statement in IPCC AR4 may lead to an underestimate of the greenhouse gas contribution to warming, because it implicitly includes the lesser known masking effect of cooling aerosols.
- Self-reported media exposure is higher for those who are skeptical of a significant human influence on climate.
In 2012, while temporarily based at the Netherlands Environmental Assessment Agency (PBL), my colleagues and I conducted a detailed survey about climate science. More than 1800 international scientists studying various aspects of climate change, including e.g. climate physics, climate impacts and mitigation, responded to the questionnaire. The main results of the survey have now been published in Environmental Science and Technology (doi: 10.1021/es501998e).
Level of consensus regarding attribution
The answers to the survey showed a wide variety of opinions, but it was clear that a large majority of climate scientists agree that anthropogenic greenhouse gases are the dominant cause of global warming. Consistent with other research, we found that the consensus is strongest for scientists with more relevant expertise and for scientists with more peer-reviewed publications. 90% of respondents with more than 10 climate-related peer-reviewed publications (about half of all respondents), agreed that anthropogenic greenhouse gases (GHG) are the dominant driver of recent global warming. This is based on two different questions, of which one was phrased in similar terms as the quintessential attribution statement in IPCC AR4 (stating that more than half of the observed warming since the 1950s is very likely caused by GHG).
Figure 1. The more publications the respondents report to have written, the more important they consider the contribution of greenhouse gases to global warming. Responses are shown as a percentage of the number of respondents (N) in each subgroup, segregated according to self-reported number of peer-reviewed publications.
Literature analyses (e.g. Cook et al., 2013; Oreskes et al., 2004) generally find a stronger consensus than opinion surveys such as ours. This is related to the stronger consensus among highly published – and arguably the most expert – climate scientists. The strength of literature surveys lies in the fact that they sample the prime locus of scientific evidence and thus they provide the most direct measure of the consilience of evidence. On the other hand, opinion surveys such as ours can achieve much more specificity about what exactly is agreed upon and where the disagreement lies. As such, these two methods for quantifying scientific consensus are complementary. Our questions possibly set a higher bar for what’s considered the consensus position than some other studies. Furthermore, contrarian viewpoints were likely overrepresented in our study compared with others.
No matter how you slice it, scientists overwhelmingly agree that recent global warming is to a great extent human caused.
Figure 2. The more publications the respondents report to have written, the more they agree with greenhouse gases being the main contributor to global warming (red bars). IPCC AR4 authors report the most agreement with GHG-driven global warming. Results are based on two different questions, one about the qualitative contribution of GHG (Q3) and one about the quantitative contribution of GHG (Q1).The latter question resulted in more “undetermined” answers (unknown, I don’t know, or other), presumably because it was more difficult to answer. Hence the percentage of consensus can best be compared by excluding these “undetermined” answers. Responses are shown as a percentage of the number of respondents in each subgroup.
The concept of ‘consensus’ has been discussed a lot lately. Whereas the presence of widespread agreement is obviously not proof of a theory being correct, it can’t be dismissed as irrelevant either: As the evidence accumulates and keeps pointing in the same general direction, the experts’ opinion will converge to reflect that, i.e. a consensus emerges. A theory either rises to the level of consensus or it is abandoned, though it may take considerable time for the scientific community to accept a theory, and even longer for the public at large.
Greenhouse warming versus aerosol cooling
By phrasing Question 1 analogously to the well-known attribution statement of AR4 we found something peculiar: Respondents who were more aware of the cooling effect of aerosols in greater numbers assessed the greenhouse gas contribution to recent warming to be larger than the observed warming (consistent with the IPCC assessments). We concluded that the AR4 attribution statement may lead people to underestimate the isolated greenhouse gas contribution. The comparable AR5 statement is an improvement in this respect.
Media exposure
Respondents were also asked about the frequency of being featured in the media regarding their views on climate change. Respondents who thought climate sensitivity was low (less than 1.75 degrees C per doubling of CO2) reported the most frequent media coverage. Likewise, those who thought greenhouse gases had only made an insignificant contribution to observed warming reported the most frequent media coverage. This shows that contrarian opinions are amplified in the media in relation to their prevalence in the scientific community. This is related to what is sometime referred to as “false balance” in media reporting and may partly explain the divergence between public and scientific opinion regarding climate change (the so-called “consensus gap”).
Figure 3. The most frequent media coverage is reported by respondents who estimated climate sensitivity to be lowest (<1.75 degrees for a doubling of CO2 concentration). Respondents are grouped according to their estimate of climate sensitivity (colored bars). Responses are shown as a percentage of the number of people (N) in each climate sensitivity range.
Survey respondents
Respondents were selected based on a few criteria: Having authored articles with the key words ‘global warming’ and/or ‘global climate change’, covering the 1991–2011 period via the Web of Science. This is the same database used by Cook et al in their recent ERL study (PS: John Cook is co-author on this current study as well). Respondents were also selected based on inclusion in the climate scientist database assembled by Jim Prall, as well as by surveying the recent climate science literature. Prall’s database includes signatories of public statements disapproving of mainstream climate science. They were included in our survey to ensure that the main criticisms of climate science would be included. This last group amounts to less than 5% of the total number of respondents, about half of whom only published in the gray literature on climate change.
Survey questions
Detailed questions were posed about a variety of physical climate science issues, which are discussed in the public debate about climate change. Answer options reflected a variety of viewpoints, all of which were phrased as specific and neutral as possible. Before executing the survey, questions and answers (pdf) were reviewed by physical and social scientists and climate change public commentators with a wide range of opinions (see acknowledgements for a list of names), to minimize the chance of bias.
Comments on the survey by respondents varied: some said it was slanted towards the ‘alarmist’ side (“Obviously these questions were posed by warmists”), but more respondents commented that they thought it was slanted towards the ‘skeptical’ side (“I suspect this survey comes from the denial lobby”).
——————-
Reference: Bart Verheggen, Bart Strengers, John Cook, Rob van Dorland, Kees Vringer, Jeroen Peters, Hans Visser, and Leo Meyer, Scientists’ Views about Attribution of Global Warming, Environmental Science and Technology, 2014. DOI: 10.1021/es501998e. Supporting Information available here. The article is open access.
An FAQ for this article is here.
A Dutch version of this blogpost is here.
Tags: Attribution, cliamte science, Climate Science Survey, false balance, greenhouse gases, John Cook, media exposure, PBL, public debate, scientific consensus
My view on climate change 
August 11, 2014 at 15:23
Will examine the paper in full when I can print it.
Problems with this study are two-fold:
[1] The use of Cook et al’s data as source. Cook’s search terms: ‘global warming’ and ‘global climate change’, while appearing intuitive, does not fetch relevant papers from Web of Science. Most of Cook et al’s papers are retrieved by the search phrase ‘global warming’. The authors failed to reveal this bit of information in their paper. ‘Global warming’ is a phrase used by workers in the climate impacts and mitigation fields. It stands to reason their database of email addresses is constituted predominantly by researchers working in these areas as the addresses were sourced from the papers.
Scientists and workers in these areas are not suitable targets to determine in-depth views on orthodox IPCC climate science positions on attribution. Additionally, they are unlikely to disagree with IPCC positions. It would violate assumptions of their own fields of expertise.
[2] Jim Prall has an extensive list of scientists publishing in climate fields. Prall’s list, however, includes scientists associated with the IPCC reports themselves. For instance, the list included 916 scientists who were contributors to the IPCC AR4 report. Interviewing scientists who were authors of the IPCC reports on whether they agree with IPCC formulations is circular. Dennis and Bray have previously demonstrated that IPCC authors are likely to agree with summary conclusions of reports they were authors on.
The real interesting remaining bit in the data is the fraction of respondents that were not in Cook’s ‘global warming’ retrieved authors, or not IPCC AR4 or AR5 authors. Closer examination of the paper’s data might provide useful information, on *how big this fraction really is*. Or, if the authors pursue the same data practices as one of the present papers’ authors, John Cook, we might never learn the answer to such questions.
August 11, 2014 at 16:27
Did any of the 1800 explain the Pause adequately?
August 11, 2014 at 16:35
Commenter Shub’s second complaint makes no sense. The study results isn’t “people who agree with a committee”. It’s “people who agree about a feature of the universe”. It’s totally appropriate to include people who have previously said they agree. Wtf, Shub?
August 11, 2014 at 16:42
sunshinehours1, anyone can explain the “pause”. You have to zoom in perfectly and ignore data to even think it exists. Zoom out and don’t cherry-pick data.
The problem with climate science is that too many citizens think the amount of time a sit-com last is an appropriate amount of time to judge trends of an entire planet.
August 11, 2014 at 16:47
The problems with this study are indeed two-fold:
[1] I guess, by septics standards, Bart, you shoud’ve sent the questionaire to the Oregon Petition signatories. 30 thousand highly qualified septic scientists opposing human-caused global warming. 30k is bigger then 1.8k, no? Shub wins.
[2] Any study that carries John Cooks name will never be accepted by septics. That in itself is really interesting.
Oh, and a third, that is not so much a problem as a warning: beware of Tol and his many gremlins on Twitter.
August 11, 2014 at 17:18
Well, clearly, this study was flawed because I don’t remember receiving a survey! And I’m even on Prall’s list (top 1500).
In any case, thank you for doing this study. I personally think that the Cook & Oreskes papers are examples of results that are correct (e.g., that the majority of scientists with climate expertise believe that human-induced increased GHG levels have contributed the majority of recent observed warming), but methodologies that are imperfect (e.g., a large number of the abstracts that contribute to the “97%” number from Cook et al. don’t really state anything regarding the attribution question).
Not only that, but I found the question regarding the quantitative percentage of the human contribution interesting, because I’ve long thought that the best estimate was >100% for GHG contributions to warming (I even think it is as likely as not that GHG+aerosols contribute >100% of recent warming).
-MMM
August 11, 2014 at 18:07
Co-authors?
http://www.joseduarte.com/blog/ignore-climate-consensus-studies-based-on-random-people-rating-journal-article-abstracts
August 11, 2014 at 18:54
Chad … Chad … Chad. The Pause exists. Denying it exists is so sad.
August 11, 2014 at 19:57
For Chad
Anderegg et al, of which Jim Prall was a co-author, purported to examine expertise level in researchers who agreed or disagreed with ‘the consensus’. The ‘consensus’ opinion was defined as the IPCC AR4 attribution statement. 916 authors of the IPCC AR4 WG1 were included in the study on the basis of their agreeing with the consensus opinion, i.e., the very statement chosen by the authors to represent it!
The circularly originated data has now become the source of a second study, namely the present one.
August 11, 2014 at 22:51
Nice. And quantifying “The most frequent media coverage is reported by respondents who estimated climate sensitivity to be lowest” is cute.
August 12, 2014 at 08:55
WordReference Random House Learner’s Dictionary of American English © 2014
con•sen•sus /kənˈsɛnsəs/
n., pl. -sus•es.
[countable;
often singular] unanimous judgment or belief that a group comes to after discussion: The consensus was that they should meet twice a month.
So, anything less than 100% is not consensus
Anyway, of course, “consensus” is a political term, not scientific.
August 12, 2014 at 09:43
Heber,
I can easily counter with another definition (e.g. Merriam-Webster “a general agreement about something”). The point is that in the paper is clearly defined what is meant with consensus. Also in the whole broader discussion consensus refers to broad agreement, not necessarily unanimity. There isn’t even unanimity about the near-sphericity (sp?) of the earth, so the concept of strict unanimity in the context of scientific opinion is fairly meaningless.
August 12, 2014 at 18:07
You didn’t mention what was to me the most interesting and surprising result of the study: The majority of respondents who thought that the equilibrium climate sensitivity is at the bottom of the IPCC range. You only mentioned the sensitivity as regards a much more tenuous side point: the expertise of the various opinions on it – without telling us what they said.
I don’t mean to be uncharitable, but this most interesting result of your study is also the one that skeptics will like – and that’s the one you left out.
Note also that the famous “97%” figure is not supported by your study. There is a strong consensus, but there are dissenters too, as sensible people knew all along.
The study looks really nice, by the way. Obviously I’m not in a position to have looked at the data in detail, but the questions look good and you got a good response rate from a nice sample of scientists.
August 12, 2014 at 18:31
I should correct what I said: it wasn’t a majority who said that the sensitivity is at the bottom of the IPCC range. However, the lower numbers were a very large minority (~350 compared to 450 who chose the middle of the IPCC range, and ~125 chose the higher values).
August 12, 2014 at 19:05
Taking a closer look at the climate sensitivity question: you say that “On the other hand, the skewed distribution shown in Figure 8,
with more responses for lower rather than higher values of
ECS, is different from the distribution as inferred from theory
and as assessed by the IPCC, which has a fat tail toward higher
values.” I’m not sure this is an appropriate comparison: I think there is a difference between a distribution of everyone’s best guess to a value, and the distribution of that value.
For example, harkening back to my days of roleplaying games, say I roll 3 six sided dice behind a screen. I ask people, “what’s the most likely sum of the 3 dice?” I should get, assuming a mathematically literate audience, either 10 or 11. But the actual distribution runs from 3 to 18.
So it would have been interesting to ask in addition to what the best guess of CS was, what the 95% distribution was. And I think what you’d get is a large number of long-right-tailed distributions, even if more of them have means clustered between 2 and 3. And it would be interesting to then average those distributions and see the characteristics of this new pdf.
(the Granger Morgan expert elicitation from 1995 did just that: http://stephenschneider.stanford.edu/Publications/PDF_Papers/Morgan_KeithSM.pdf – see Figure 1. Everyone assumes that Expert 5, with the ridiculously low mean and ridiculous amount of certainty, was Richard Lindzen. Also, it is a pity you didn’t cite this paper, or its 2010 sequel, http://keith.seas.harvard.edu/papers/129.Zickfeld.etal.ExpertJudgements.e.pdf, though http://link.springer.com/article/10.1007%2Fs10584-012-0620-4 suggests that there may be some issues with this approach)
-MMM
August 13, 2014 at 00:54
> Cook’s search terms: ‘global warming’ and ‘global climate change’, while appearing intuitive, does not fetch relevant papers from Web of Science.
What search terms would you use, Shub, and which operational definition of relevance do you have in mind?
August 13, 2014 at 02:44
Agreement on a subject does not indicate the agreement has reached to correct conclusion.
A very recent, within memory of some, conclusion that the continents do not drift was proven wrong once the tech became available to show that, in fact, they do.
Dr. Wegener was ridiculed for his findings when presented, but as we all now know, he was correct, the consensus was wrong.
The divergence/pause problem is a real one in regards to the presented climate models of today. The error range of TOA measurements indicate that the perceived energy imbalance is not nearly as large as some would promote.
One thing of certainty, the earth is responding in a very positive means. Vegetation has not been this healthy, food production not this large in centuries. Man kind is benefitting greatly.
August 13, 2014 at 04:18
Sigurdur that’s an utterly ridiculous claim.
A) Since Wegener was not able to come up with a plausible mechanism, it was proper for the scientific community to reject his hypothesis pending further data.
B) He was ridiculed by an number of geologists for being a meteorologist and supposedly out of his depth – but that’s how human enterprise goes, a lot of back bitting. It doesn’t change the fact of Wegener’s lack of solid evidence.
C) Wegener’s continental drift ideas remained alive and inspired subsequent scientists to discover and collect the missing data needed to make an overwhelming case with the power to overturn established consensus. AS THEY DID and AS IT WAS.
~ ~ ~
What divergence problem?
You seem to think climate models are supposed to accurately mimic our weather. Climate models are tools to help develop our understanding! Expecting perfect replication of weather patterns is an act of Willful Ignorance and disregard for the models roll in scientific study.
What pause are you talking about? Have you no conception of what’s been happening at the north and south poles???
WHERE IS THIS PAUSE EVIDENT? Please answer that.
Or do you think this climate transition is all supposed to play out along the timescale of quarterly profit reports?
~ ~ ~
Love your little “Food production has not been this healthy in centuries” stop fixating on the Rear View Mirror.
It’s the past decade and future you should be thinking about.
http://www.epa.gov/climatechange/impacts-adaptation/agriculture.html
http://www.theguardian.com/environment/2014/mar/31/climate-change-food-supply-un
~ ~ ~
Sigurdur , your Republican/Libertarian fabricated arguments demand a total rejection of rational evidence assessment and learning – shame on you!
August 14, 2014 at 17:39
Willard asks: What search terms would you use, Shub, and which operational definition of relevance do you have in mind?
Eli answers: “Richard Tol wrote 253 papers”
As to relevance anything that Richard’s gremlins let through.
August 14, 2014 at 23:11
“What search terms would you use, Shub, and which operational definition of relevance do you have in mind?”
The question that should be asked is: How have the authors tested the suitability of abstracts retrieved by their chosen keywords for the purpose of determining ‘consensus’? If so, what method did they use?
The answer is: they did not test this.
August 15, 2014 at 04:40
Shub Niggurath Says: August 14, 2014 at 23:11
The question that should be asked is: How have the authors tested the suitability of abstracts retrieved by their chosen keywords for the purpose of determining ‘consensus’? If so, what method did they use?
~ ~ ~ ~ ~ ~ ~
> Cook’s search terms:
‘global warming’ and
‘global climate change’,
while appearing intuitive, does not fetch relevant papers from Web of Science.
Why?
Researchers assign tags to their papers.
Your complaint makes no sense.
Can you explain?
Would “Anthropogenic Global Warming” been more suitable?
Why?
What other terms would you suggest?
Why do you believe that would have made a difference in the results of the study?
~ ~ ~
And what about the gorilla-in-the-room you are ignoring – that being the substance of the message the sum total of those reports are revealing? {Why do you folks never show any interest in that?}
August 15, 2014 at 13:57
citizenschallenge
‘Global warming’ is a phrase used by workers in the climate impacts and mitigation fields. The term is colloquial and used in a non-technical sense. Researchers working on climate impacts assign this as keyword to their abstracts on the basis of their assessment of relevance to the topic of global warming. It is not reflective of their expertise or relevance in addressing questions of attribution. Additionally, what this captures is a bandwagon effect.
Search for “global warming” in WoS with settings used by Cook et al yields 9,466 abstracts, i.e., ~77% of the total 12,332 returned when using both terms. In Cook’s data, 77% of abstracts (9166/11944) are from climate impacts and mitigation categories.
The above shows the Cook et al author database is extremely likely constituted by a majority of academics whose expertise does not lie in climate physics, climate modeling or attribution methodology.
Ari Jokimakki of Skepticalscience undertook an interesting method to verify the composition of retrieved abstracts. The results are available on their website. They agree with the above.
August 15, 2014 at 17:07
I’m not finding the article at SkS by Ari that you refer to. Have a link?
~ ~ ~
Beyond that I’m not figuring out what your point is about the study or more importantly it’s conclusion.
It’s like fighting over loose change or spending an hour figuring the tip you’re going to leave for an expensive dinner.
We have only so much time and energy available, that’s why I’m constantly flabbergasted at the amount of energy and time spent arguing about what seems like window dressing, all the while ignoring the substance of the information we are being given.
Or the fact that Cook’s study is only one of many that have arrived at very similar conclusions.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
FAQ for the article “Scientists’ Views about Attribution of Global Warming”
By Bart Verheggen
Conclusions
4. How does this study compare to the often-quoted 97% consensus?
Our results are consistent with similar studies, which all find high levels of consensus among scientists, especially among scientists who publish more often in the peer-reviewed climate literature.
Cook et al. (2013) found that 97% of papers that characterized the cause of recent warming indicated that it is due to human activities. (John Cook, the lead author of that analysis, is co-author on this current article.) Similarly, a randomized literature review found zero papers that called human-induced climate change into question (Oreskes, 2004).
Other studies surveyed scientists themselves. For instance, Doran and Kendall-Zimmermann (2009) found lower levels of consensus for a wider group of earth scientists (82% consensus) as compared to actively publishing climatologists (97% consensus) on the question of whether or not human activity is a “significant contributor” to climate change. Our results are also in line with those of e.g. Bray and von Storch (2008) and Lichter (2007).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Shub Niggurath Says:
August 14, 2014 at 23:11
“What search terms would you use, Shub, and which operational definition of relevance do you have in mind?”
The question that should be asked is: How have the authors tested the suitability of abstracts retrieved by their chosen keywords for the purpose of determining ‘consensus’? If so, what method did they use?
The answer is: they did not test this.
~ ~ ~
Shub I’m wondering if you could give an example of how you think they should have approached this. Is it other search terms they should have tested… how would you have done it … can you explain?
August 16, 2014 at 18:12
The list of questions asked mentioned one about the respondents level of concern about catastrophic effects of climate change. But I couldn’t find that in the article itself. Anyone see what were the percentages of the responses to that question?
See the list of survey questions here:
Click to access Climate_Science_Survey_Questions_PBL_2012.pdf
It appears as question 12:
“Concern.
How concerned are you about climate change as a long-term global problem?”
However, I wasn’t able to find the question addressed in the article itself.
Bob Clark
August 16, 2014 at 19:41
Robert Clark, that’s correct. The answers to this Q and to many others were not included in this article.
August 17, 2014 at 04:16
I would like to know what the breakdown of the responses on that question is. My point is the key point that should be addressed is not just whether humans caused most global warming but also whether it is expected to cause catastrophic effects on a relatively short time scale, say, less than 100 years.
Bob Clark
August 17, 2014 at 14:45
Robert,
I’ll get back to that question in due time.
August 18, 2014 at 00:28
Robert Clark Says:
August 17, 2014 at 04:16
. . . whether it is expected to cause catastrophic effects on a relatively short time scale, say, less than 100 years.
~ ~ ~ ~ ~ ~ ~
Considering that question, it seems to me like our society has gotten to be adept at forgetting the environmentally destructive trends of this past century.
As for the potential for catastrophic effects, perhaps it’s time for someone to define exactly how catastrophic they want these effects to become before taking notice?
I say this because it seems that these past years have brought genuine warming climate related catastrophes to many people, after all, every weather pattern these days is born within this warming climate system of ours!
… and it’s guaranteed that weather trends will continuing destabilizing and having more energy and moisture to inject into our global heat distribution engine’s weather systems.
What are we in for in the next decades, let alone century?
Here are some of the cumulative impacts to figure into your equation:
Sea level rise risk
http://www.cbc.ca/news/canada/british-columbia/sea-level-rise-underestimated-say-b-c-scientists-1.1275818
http://citizenschallenge.blogspot.com/2014/07/port-authorities-rising-sea-levels.html
http://www.centerforoceansolutions.org/montereybay/sea-level-rise/local-variability
~ ~ ~
Ocean acidification risk
http://citizenschallenge.blogspot.com/2014/04/ocean-acidification-already-eating-away.html
Craig Welch | Seattle Times | April 30, 2014
Click to access frontpage.pdf
~ ~ ~
Forest health
http://www.scientificamerican.com/article/out-on-a-limb-global-warm/
http://www.smh.com.au/environment/climate-change/forest-fires-worsen-global-warming-to-thaw-greenlands-ice-20140520-zri0f.html
and why it matters
http://www.huffingtonpost.com/2012/09/09/climate-change-stress-killing-forests_n_1868678.html
~ ~ ~
Farming
Click to access 2012EO200003_nws.pdf
Click to access climatechange4.pdf
~ ~ ~
Energy supply
http://www.ucsusa.org/global_warming/science_and_impacts/impacts/effects-of-climate-change-risks-on-our-electricity-system.html
http://www.wri.org/blog/2013/02/how-climate-change-impacts-america’s-energy-infrastructure
~ ~ ~
And these are just some of the more obvious ones.
Seems to me a null hypothesis could be suggested:
How could such historic and unchecked trends not lead to a series of unprecedented catastrophes?
August 18, 2014 at 01:20
Oh boy, what a story to I bump into after posting the above.
here’s another one for that “potential CAGW” equation
Tundra melt, methane
The Really Scary Thing About Those Jaw-Dropping Siberian Craters
BY ARI PHILLIPS POSTED ON AUGUST 1, 2014 AT 10:39 AM
http://thinkprogress.org/climate/2014/08/01/3466466/siberian-craters-permafrost-climate-change/
~ ~ ~ ~ ~ ~ ~
While I hear others talk about centuries, it sure seems to me we are talking decades and damned am I glad I’m about to turn 60 rather than 16… and oh what a hideous indictment of my generations.
We’ve know since the 60/70s what we were playing with.
By the 80/90s we had the evidence.
Here we are 2014 and nothing of substance has happened except exponentially rising consumption and rampant biosphere pollution and destruction and calls for more study.
I don’t see how any new fresh study can tell us anything significantly new, but another new study will be called for so we can be absolutely positively certain yet again that it can be worse than the previous study warned us.
And they call me the fool, lordie lord.
{FYI. Rob my whining is not addressed to you, it towards that PR machine that’s responsible for this insane state of affairs.}
August 18, 2014 at 10:42
Please keep comments on this post on-topic, i.e. related to this survey.
There’s an open thread for other climate related discussions.
August 18, 2014 at 12:52
“Shub I’m wondering if you could give an example of how you think they should have approached this. Is it other search terms they should have tested… how would you have done it … can you explain?”
–
citizens, The answer would concern issues with Cook et al, which is not the topic of discussion here. Briefly however, the answer is simple: while reading abstracts to survey the scientific literature seems an attractive thing to do, the exercise is fraught with methodological and sampling issues, some of which are insurmountable. The keyword question is elaborated above. Secondly abstract text, retrieved by whatever method, does not contain the kind of consensus information required for objective raters to carry out classification. Thirdly, rater biases impact classification.
The best option would be to perform direct surveying (as scientists).
Direct surveying requires sound sampling strategies as well. With lack of data, it is impossible to assess sampling for this paper.
August 19, 2014 at 20:40
The Goat’s objections appear to invalidate our entire approach to pursuing scientific progress. Publications by themselves are useless; only personal, direct interviews with researchers are sufficiently reliable as to allow citation of past work as supporting material for new research.
Unfortunately the Goat’s brilliant epiphany has thereby destroyed the notion of foundationalism, thus exposing us as knowing virtually nothing with reasonable confidence. For example, we cannot speak to Newton and obtain confirmation of his basic insights on the three laws of motion, thus our confidence in classical physics is misplaced, etc. This is a pretty embarrassing situation to say the least.
On the face it of the Goat’s claim seems somewhat unreasonable.
August 31, 2014 at 18:32
While you can criticize Cook et. al along some dimensions, The criticism that the large number of articles with no specific attribution invalidates it is silly.
Here’s a thought experiment:
1. Take a random collection of published articles in biology journals.
2. Count the number which actually dispute or affirm evolution. If this number is over 10% I’d be very surprised. If the number denying evolution is over .5% I’d be extremely surprised.
3. Now argue that since 90% of articles made no specific attribution, there is limited consensus among biologists over evolution.
That would be a very silly argument
August 31, 2014 at 23:59
JGarland wrote: “While you can criticize Cook et. al along some dimensions, The criticism that the large number of articles with no specific attribution invalidates it is silly.”
Your thought experiment only shows that those ambiguous papers will not prove a limited consensus. But since those papers provide no valuable information at all, they cannot be used to support it either. How is a study valid if the majority of your data is worthless?
Actually it’s worse than that
http://www.joseduarte.com/blog/cooking-stove-use-housing-associations-white-males-and-the-97
September 4, 2014 at 21:06
amatterofmarching jumps off the same cliff as Jose Duarte, not understanding what “consensus” means.
A consensus does *not* mean that the majority of papers explicitly determine that the dominant theory is indeed correct. A consensus means that the majority of scientists consider it the most likely theory, and either explicitly or implicitly apply it to their work. Which directly explains why a lot of the papers that Duarte complains about are included in the count.
If you look for papers on evolution, the vast majority don’t show how it works, but just apply it to their work, showing they accept it as the prime explanatory theory. The same goes for HIV causing AIDS; the theory of relativity; anthropogenic global warming.
September 5, 2014 at 22:59
Marco, papers that provide no information one way or the other cannot be used to prove a consensus or a lack of consunsus. Anything beyond that is subjective, silence is not the same as agreement (or disagreement).
It might have helped to actually read Duarte’s article.
“The above papers have nothing to do, epistemologically, with the scientific consensus on global warming. The consensus only pertains to climate science, to those scientists who actually study and investigate climate. To include those papers was either a ridiculous error or fraud. I didn’t expect this — I expected general bias in rating climate papers. I never imagined they’d include surveys of the public, psychology papers, and marketing studies. In retrospect, this was entirely predictable given that the researchers are a bunch of militant anti-science political activists.”
He has written more about the methodology. Worth looking at:
http://www.joseduarte.com/blog/ignore-climate-consensus-studies-based-on-random-people-rating-journal-article-abstracts
And more about other studies from the same group of individuals:
http://www.joseduarte.com/blog/lewandowsky-fraud
http://www.joseduarte.com/blog/more-fraud
BTW, Duarte agrees with the consensus
September 6, 2014 at 00:08
From Duarte’s website: “I’m a PhD candidate in Social Psychology at Arizona State University, a few weeks counting down to my defense. My research focuses on envy, how people respond to others’ achievements in general (beyond envy), and different forms of self-esteem (secure vs. fragile, narcissistic, authentic, etc.) I have equally strong interests in methodology, especially validity, bias, and philosophy of science.”
How does put him in a qualifying position to pass judgement – freely labeling people: ”the researchers are a bunch of militant anti-science political activists.”
~~~~~~~~~~~~~~~~~~~~~~~~~~~
So what’s Duarte’s ax to grind? Or perhaps just another opportunist taking advantage of the hot new marketing niche ??
September 6, 2014 at 11:15
centr5ral, I’ll only take one example that is clear and obvious evidence Duarte is an idiot:
“Most of these studies use political activists as the raters, activists who desired a specific outcome for the studies (to report the highest consensus figure possible)”.
The evidence he provides for the rates being political activists who desired a specific outcome is…drum roll…because he says so. He provides no evidence whatsoever for that claim.
It gets worse, however. The paper clearly and openly acknowledges potential bias of the raters. They tested that potential bias by comparing with the author ratings. It probably was too inconvenient for Duarte that this showed the raters had been *conservative* in their ratings compared to what the authors said, so let’s not mention that, eh?
His continuous cries of “fraud” are perhaps more telling about his own frame of mind than reality.
cc, I think Duarte has become wedded to his beliefs that it’s not that conservatives are wrong, it’s that they are being discriminated against in academia.
September 1, 2015 at 00:25
I personally believe that more than 50% of 20th-century warming can be attributed to anthropogenic forcing, but the controversy over this meaningless political statement is one of the stupidest in climate science. It is both meaningless and consistent with ethical science.
Suppose exactly 50% of 20th-century warming could be attributed to anthropogenic forcing, leaving unforced variability (and perhaps solar and volcanos) as the explanation for the remaining 50%. Anthropogenic warming this low plus known anthropogenic forcing produces estimates of ECS and TCR between 0.6 and 1.0 degC! Yes, I’m 95% confident that ECS and TCR are greater than 0.6-1.0 degC. So what? These values won’t produce catastrophic climate change and certainly don’t justify emissions reductions. You need to blame ALL of 20th century warming on man and 150-200% of 20th-century warming on GHGs before the future looks catastrophic. The forcing from GHGs today is nearly equivalent to a doubling of CO2 right now.
As Stephen Schneider points out, ethical science means including all of the if, ands, buts and caveats. What caveats should have been attached to the attribution statement?
How about starting with: “If climate models accurately reproduce unforced variability in climate, then we conclude …” What evidence do we have that climate models show skill in reproducing unforced variability in climate. such as: ENSO (poor), AMO (negligible?), the recent hiatus (poor), warm period around 1940 (none), hindcasting decadal variability (CMIP5, none), the LIA (poor), the MWP (none), orbital forcing producing and ending ice ages (none).
“If climate models accurately reproduce unforced variability AND climate sensitivity, then we conclude …” Do climate models get ECS right? Most models have ECSs around 3 degC. The IPCC gives a 15-85% confidence interval of 1.5 to 4.5 degC for ECS. The confidence intervals for attribution statements based on GCMs are too narrow.
September 1, 2015 at 08:57
Mostly agree with you, except with that “more than 50% can be attributed”; paleoclimate records make me think that that attribution is more likely just 0,1% due to land use.
But the most important thing to me is your real scientific attitude, or should I say your not too common “common sense”?
September 1, 2015 at 11:37
Please keep the discussion focused on the survey results.
Thanks!